F'IRETREE (MYRICA FAYA) DISTRIBUTION IN HAWAICI

Louis D. Whiteaker and Donald E. Gardner

ABSTRACT

The primary objective of this study was to map the distribution and abundance of firetree (Mynca faya) in the state of Hawai'i. Reconnaissance data were used to map 85,912 a (34,365 ha) of infestation throughout the State. Infestations of 72,265 a (28,906 ha) occur on the island of Hawai'i, 4,770 a (1,908 ha) on Maui, 2,518 a (1,007 ha) on Lha'i, 435 a (174 ha) on O'ahu, and 5,925 a (2,370 ha) on Kaua'i. Distribution ranges in elevation from as low as 1,400 ft (425 m) on Liina'i to as high as 6,400 ft (1,940 m) on the slopes of Haleakall on Maui. Firetree occurs on recent, thin ash over pihoehoe lava as well as on deep, well-developed silty clay loam soil. It is found in montane rain forest habitats and in dry scrub (marginal through submontane seasonal) forest. The distribution patterns of firetree suggest that this species - has not yet reached the limits of its potential distribution in Hawai'i. Observations on the phenology of firetree were made at two-week intervals from March 1983 through March 1985. Data were taken at three sites and were related to the data from nearby weather stations. Information on flowering, fruiting (mature and immature), leaf flushing, leaf fall, and fruit drop was collected. Analysis of the phenologic patterns observed showed flowering, fruiting, and fruit drop to be endogenously controlled, while leaf flushing and leaf fall were influenced by environmental variables.

INTRODUCTION

Fire or faya tree (Myrica faya, Myricaceae) is a native of the Azores, Madeira, and the Canary Islands in the Atlantic Ocean. It was introduced to Hawai'i in the late 1800s by Portuguese immigrants, resumably as an ornamental (Fosberg 1937). Because it forms multiple

%ranches near the base of the main stem (Smathers and Gardner 1979), firetree has often been described as a shrub or small tree reaching heights of only 13 to 20 ft (4-6 m) (Fosberg 1937; Neal 1965; Hasselwood and Motter 1983). However, in the H&n&ua region of the island of Hawai'i, firetree

Whiteaker and Gardner \ Firehe Distribution in Hawai'i 226

grows to over 50 ft (16 m) high, forrnin dense cano ies with an understory devoid of other plant life (Smathers and sardner 19797.

Firetree has narrow, pointed, smooth, shiny, dark green leaves with entire or toothed margins (Neal 1965; Hasselwood and Motter 1983). Although firetree has a strong tendency toward dioecism, a few staminate flowers are found on pistillate plants and a few pistillate flowers are found on staminate plants (Gardner 1985). Staminate flowers with four stamens each are borne on small catkins. Three pistillate flowers that may be joined and accompanied by one bract are also grouped in small catkins. Fruits are small, edible drupes that form dense clusters changing from green through red to purple when ripe (Fosberg 1937; Lawrence 1951; Neal 1965).

The purpose of this stud was to map the distribution and abundance of firetree and to observe t g e phenological cycles of this species in the state of Hawai'i. Distribution studies will provide a database for location of all infestations, both for implementation of a biological control pro am, should an appropriate agent be found, and for future studies of t i? e spread of this species in Hawai'i. The phenology data will help to determine the timing of control programs.

HISTORY OF SPREAD AND CONTROL EFFORTS

The Hawaiian Sugar Planters' Association obtained seeds of firetree from a Portuguese farmer on the island of Hawai'i for use in reforestation attempts (Fosberg 1937). Plantings were made on the islands of Kaua'i, O'ahu, and Hawai'i (Skolmen 1979), most of them in the 1920s. The aggressive, noxious character of this s ecies was soon noted, along with its spread to Maui (Fosberg 1937). &e continued spread of firetree led the Territorial Board of Agriculture and Forest to begin attempts to eradicate it in 1944 (Neal 1965). Managers of S X ipman Estate observed firetree spreadin in the Volcano area on the island of Hawai'i in the mid-1940s. In t f e mid-1960s, Shipman attempted to control the species there (T. Lindsey, pers. comm.). It was declared noxious for state land leases because it spreads rapidly and forms a dense cover that crowds out desirable species (Hosaka 1945; Haselwood and Motter 1983). Firetree has continued to spread despite efforts by the State and the National Park Service, and it now occurs on all major Hawaiian Islands except Kaho'olawe, Moloka'i, and Ni'ihau. Estimates of the total infested area have increased over the years: 8,200 a (3,280 ha) 1954), 21,375 a (8,550 ha) (Anonymous 1962), 40,000 a (16,000 and Null 1970), and the most recent estimate, showed to be supporting firetree po ulations of various densities Hawai'i contained 50,000 a &000 ha); Maui, 3,000 a (1,200 ha); O'ahu, 100 a (40 ha); Kaua'i, 325 a (130 ha); and bna'i , 512 a (205 ha).

Firetree was thought to occur where average annual rainfall is 35 in. (900 mm) or more in mesic low-elevation forest and wet low- and hi h- e l eva t i on f o r e s t [ vege t a t i on zones C1, C Dl, and % of Ripperton and Hosaka (1942)l (Hosaka and Thistle 1 9 5 ~ ~ n o n ~ m o u s 1962f

Khiteaker and Gardner \ Firehe Distibution in HawaiS' 227

Within these broad limits, however, the tree can adapt to a wide range of habitats. Clarke 1978 described the distribution of firetree in Hawaii Volcanoes Nation Par 1; as a horizontal band between 2,200 and 4,000 ft (665-1,210 m) elevation, including 11 soil es and 15 of the vegetation 9 units described by Mueller-Dombois and Fos erg (1974). Average annual rainfall within this distribution is from 50 in. (1,270 mm) to over 100 in. 2,540 mm), and average annual tem erature ranges from 60 to 72 F (22 C) K Clarke 1978). The wide range of abitats where firetree has been found

suggests that it has not yet occupied its full potential range.

An estimated 15,000 a (6,000 ha) of infestation along the Hamakua coast on the island of Hawai'i has been classified as "hea " (Watanabe 1982). 7 In this region firetree grows over 50 ft (16 m) t a l and forms dense, interlockin canopies with no understory (Smathers and Gardner 1979). The P absence o other plant species under the canopy may be due partially to

shading, . However, allelopathic activity has been reported for a closely relate species (M. cerifera) from the southeastern United Stares (Dunevitz and Ewel 1981). Thus, the lack of understow in firetree stands hay be due not only to'canopy -shading, but also to dlelopathic activity (Smith 1985).

Current evidence suggests that the seeds of firetree are dispersed primarily by birds (Smathers and Gardner 1979; LaRosa et al. 1985). The extensive, uniform distribution of the tree in remote areas and the close spatial association of this species with other tree species, especially 'ohi'a (Metrosideros polymorpha), suggest dispersal of seeds by birds. Birds commonly associated with dispersal include the Japanese white-eye (Zosterops japonica), the common myna (Acridotheres tristis), the red-billed 'leiothrix (Leiothrix lutea), and the 'oma'o (Phaeomk obsnznrs) (Clarke 1978; Smathers and Gardner 1979; Gardner and Davis 1982; LaRosa et aL 1985). Low germination rates of firetree seeds collected in the field, contrasted with copious seed production and ra id dispersal of the species, has led to the hpothesis that scarification gom bud ingestion greatly improves germination rates (Clarke 1978). This idea was supported by experiments that showed increased amounts and rates of germination due to mechanical or chemical scarification of seeds of M. cerifera (Ewel et al. 1982). However, germination tests of firetree in Hawai'i showed no significant difference m amounts or rates of germination between seeds that passed through captive birds and apparently mature, viable seeds collected in the field from trees (LaRosa et d 1985). Feral pigs Sus scrofa) have also been identified as possible dispersal agents ! or seeds of firetree. Firetree seedlings have been observed areas in Hawaii analyses of feral pigs in firetree seeds during certain times of

Firetree control efforts have been implemented primarily by the state of Hawai'i and the National Park Service. The effort by the State has been conducted over the last 25 years, but variations in availabili of funding Y and manpower have resulted in fluctuations in intensity o this effort. Herbicides are the primary control agents. Of the various herbicides used

Whiteaker and Gardner \ Firehe Distribution in Hawai'i 228

by the State, Tordon 22K (picloram) has been the most effective, producin a com lete canopy kill and a 99% control of resproutin (Walters and Nu 1 i! f f 1970; mathers and Gardner 1979; R. Karni, pers. comm. . In Hawaii Volca- noes National Park, basal bark application of a 4% solution of KURON silvex) in diesel oil was effective, and a method to introduce ROUNDUP glyphosate) directly into the vascular tissue of firetree through a cut

branch was su ested for remote areas and less-than-ideal weather conditions (Gar CY ner and Kageler 1982 In some areas, pasturelands have been cleared of invading firetree wit b. bulldozers by private landowners. However, follow-up momtoring and/or herbicide treatment of stumps and brush piles are needed with mechanical methods to control resprouting (R. Karni, pers. comm.).

F. Bianchi, an exploratory entomologist from Hawai'i, was sent by Hawaiian Sugar Planters' Association to the native habitats of firetree in 1955 to search for potential biological control agents. A fungus disease caused by Dothiorella berengeriana initial1 appeared promising but was later rejected because tests conducted in 6ortugal showed a lack of host specificity (Gardner and Davis 1982; Gardner 1984). Other disease-causing organisms were observed on this trip, but none have been sufficiently tested for use as control agents. Krauss (1964) summarized his and others' previous observations of insects associated with firetree and related species. However, insects collected during exploratory efforts either proved ineffective in tests of control or failed to propagate. Krauss (1964) also discussed other apparently pathogenic organisms on Myrica spp., but no attempt has been made to test these as control agents in Hawai'i (Gardner and Davis 1982). Observations from another exploratory trip indicated that firetree is generally abundant and healthy in its native range and that biological agents m these areas are not important factors in limiting the growth, reproduction, or distribution of this species. However, at least two diseases and two insects were found to merit further investigation; these should be the subjects of current research in a biological control program for firetree in Hawai'i (Gardner 1984; Hodges and Gardner 1985).

DATA COLLECTION

Maps of the distribution of firetree in Hawai'i presented herein (Figs. 1-5) are based on reconnaissance data. Reconnaissance was done in or near areas where firetree was reported by knowledgeable individuals. Final drafts of the distributiori maps were plotted on U.S. Geological Survey topographic maps (Whiteaker and Gardner 1985). Phenolop data (flowering, immature fruiting, mature fruiting, leaf flushing, lea fall, and fruit drop) were collected at three sites on the island of Hawai'i (Whiteaker and Gardner 1987).

Whiteaker m d Gardner \ Firetree Dktibution in HawaiS

RESULTS

Distribution Island of Hawai'i. Firetree was found between Laupahoehoe and

Honoka'a (Hamaha District) from 2,000 to 4,400 ft (600-1,335 m) elevation, with 29,390 a (11,755 ha) infested (Whiteaker and Gardner 1985). Much of this area is ranchland where firetree has been cleared from pasture areas, but extensive areas are still covered with heavy or moderate density stands of large (65 ft or 20 m tall) trees.

Another major infestation on Hawai'i Island is in the vicinity of Volcano Village, including a large portion of Hawaii Volcanoes National Park. The distribution ranges between 1,800 and 4,000 ft (545-1,210 m) elevation and covers 41,500 a (16,600 ha). The Volcano area has a wide variety of habitats, including montane 'ohi'a rain forest near Volcano Village and the Kilauea summit area, submontane seasonal forest near ' h a h o u Ranch and Hilina Pali Road, montane seasonal forest converted to asture on Keauhou Ranch adjacent to the Mauna Loa Stri Road area of the

7 rP bark, and d scrubland with scattered trees in the a'ii Desert. The vegetation, cirnate, and substrates of the wide variety of habitats in which firetree is found within the Park have been described in detail by Doty and Mueller-Dombois (1966) and Mueller-Dombois and Fosberg (1974).

An infestation of firetree occurs on the Kapapala Ranch (Ka'u District adjacent to the Ainapo Cabin between 3,500 and 3,760 ft (1,060-1,140 m

P elevation. The infestation occu ies 80 a (32 ha). This ranchland consists of pasture mixed with stands o alien shrubs and trees on stee er slopes. Firetree has also been reported in the Kiolaka'a-Kea'a Homestea I s Addition portion of the Ka'u Forest Rese'rve at 2,300 ft (720 m) elevation.

area (L.W. Cuddihy and S.J. Anderson, pers. comm.). A plant was recorded on a Hawaii State Division of Forestry survey of t is

Firetree occurs on the slopes of Hualalai (North Kona District) between 4,850 and 6,040 ft (1,470-1,830 m) elevation. The infestation covers 1,300 a (520 ha). This population is in a remote area on privately owned ranchland (R. Kami, pers. comm.).

The total area of firetree distribution on the island of Hawai'i is 72,265 a (28,906 ha) (Fig. I), including 30,500 a (12,200 ha) within Hawaii Volcanoes National Park.

Island of Maui. Firetree is found in the Kula area on the western slopes of Haleakala between 3,200 and 6,400 ft (970-1,940 m) elevation (Fig. 2). The area consists of ranchland and small, private residential parcels of land. Total area of firetree distribution is 4,770 a

The distribution may be limited by the efforts of Haleakala Ranch, (Igo8 hab. w ich controls weed species. Only one individual firetree along the Haleakala Highway, at 4,250 ft (1,290 m) elevation, and a small stand in Hapapa Gulch were noted within the boundaries of Haleakala Ranch (Fig. 2). However, these individuals indicate that the ranch is suitable habitat for an infestation of this species. The Hawaii Department of Agriculture attempted to control firetree on Maui in the mid-1970s, but logistical

Whiteaker and Gatdner \ Firetree Dishibution in HawaiS

M AUN A LOA

HIGH DENSITY

Figure 1. Distribution of fuetree on the island of Hawai'i.

Figure 2. Distribution of firetree on the island of Maui.

Whiteaker and Garher \ Firetree Dishibution in Hawaif 231

difficulties related to the large number of small, privately owned parcels of land led to abandonment of the effort (E. Tamura, pers. comm.).

Island of Liina'i. Firetree occurs in the mountainous portion, or Liina'i Hale area, of the Island (Fig. 3). The distribution ranges from 1,400 ft (425 m) elevation in protected gullies to the summit of the Island at 3,370 ft (1,020 m). The total area occupied is 2,518 a (1,020 ha). This range of firetree on Ena ' i is rough, mountainous terrain consisting of narrow ridges and steep slopes.

Island of O'ahu. Firetree is found in the Wai'anae mountains between 2,000 and 3,127 ft (610-948 m) elevation (Fig. 4). The center of distribution is at the southern end of the Wai'anae mountains between Mauna Kapu and Pu'ukaua in the Honouliuli Forest Reserve, where it was planted by the Territory of Hawaii for reforestation (Skolmen 1979). A smaller, noncontiguous population occurs to the northwest at Pu'u Kawiwi, and to the north at Pu'u Hapapa (Fig. 4). The total area occupied is 435 a (174 ha). As on the island of L&na'i, the range is rough, mountainous terrain with narrow ridges and steep slopes.

Firetree was planted in several other areas on O'ahu, including sites in the Ko'olau mountains (Skolmen 1979). However, populations in these locations have not been observed by persons famihar with these areas (J. Obata, pers. comm.), and it is assumed that plantings were at elevations not suited for reproduction.

Island of Kaua'i. The major infestation of firetree is found just west and north of Waimea Can on, mostly within Waimea Canyon and Koke'e r State Parks. A separate popu ation just east of Waimea Canyon is centered around Wai'alae Cabin on Wai'alae Stream in the Na Pali-Kona Forest Reserve and is partially within the Alaka'i Wilderness Preserve. Firetree is also found on the northwestern slopes of the Island below the mesic native forest and/or forestry plantin s (Fig. 5). These opulations occur between 1,800 ft (545 m) on the nort f western slopes an f 4,200 ft (1,270 m) within Koke'e State Park. The total area occu ied is 5,925 a (2,370 ha). On Kaua'i, firetree occurs in a variety of ha g itats, from montane rain forest with little topographic relief to sparsely vegetated, dry, eroded, steep slopes.

All Islands. Total area of firetree distribution in Hawai'i is 85,912 a (34,365 ha). More detailed maps and descriptions of the distribution of firetree in Hawai'i are presented in a separate report (Whiteaker and Gardner 1985).

Phenology Krauss (1964) reported that firetree in its native habitat showed

abundant male flowers and a few small fruits in June; in July, most male flowers were dry, and man green (and a few purple) fruits were seen. r Krauss observed all stages o fruiting in Se tember, and by November many ripe and fallen fruits were reported. 8 ardner (1985) reported various

Whiteaker and Gardrter \ Firelwe Distibution in Hawai'i

Figure 3. Distribution of firetree on the island of h a ' i .

N

0 - 0 15 krn

Figure 4. Distribution of firetree on the island of O'ahu.

Whiteaker and Gardner \ Firetree Distribution in Hawai'i

WAIALEALE

MODERATE DENSllY

k-i 0 5 miles

Figure 5. Distribution of firetree on the island of Kaua'i.

stages of flowering with some immature fruit development during a visit to the native habitats of firetree in April and May.

In this study, flowering, immature fruiting, mature fruiting, and fruit drop showed similar patterns at all three study sites on Hawai'i Island (Hilina Pali and Byron Led e in Hawaii Volcanoes National Park, and the Eldon English property in &lcano). Flowering showed maximum activity in June and &mum activity in December and January in the two years observed. Peaks of immature fruiting followed in August and September, with minimum activity in February and March. Maximum mature fruiting was observed in November, and minimum mature fruiting was seen in May. Fruit drop followed a similar pattern to mature fruiting in the one year observed. Leaf flushing showed similar patterns at two of the sites, with maximum activity in June. At the third site, leaf flushing also had eaks in June, but the pattern over the entire observation period d' & ered somewhat from the other two sites. Two sites also had similar patterns in the amounts of leaf fall, but the amounts of leaf fall at the third site showed a different pattern. A more extensive presentation and discussion of these data, including graphs, appears in a separate report (Whiteaker and Gardner 1987).

Whiteaker and Gardner \ Firehe Distribution in Hawai'i

DISCUSSION

Distribution The distribution of firetree described here indicates that this alien

species is adapted to a wide variety of habitats in Hawai'i. It occurs on recent, thin ash over pahoehoe lava on Hawai'i Island and on soil classified by Foote et al. (1972) as deep, well-developed silty clay loam on Kaua'i and Hawai'i. It occurs in montane rain forest habitats on Hawai'i and Kaua'i islands and in dry scrub marginal to submontane seasonal forest on Kaua'i and Hawai'i.

It f ows as low as 1,400 ft (425 m) elevation on Liina'i and as high as 6,4 0 ft (1,940 m) on Maui. It is found on nearly vertical slo es on Liina'i, in volcanic craters in Hawaii Volcanoes National Par t: , and on nearly level terrain or gentle slopes in Koke'e State Park, Kaua'i. These extremes of several habitat factors in areas where firetree occurs suggest that the species may be able to occupy sites that are intermediate along these habitat factor gradients where it does not now occur. Competition from other plants may also be a factor on some sites, and some small, outlying opulations may have been missed during the survey. The distribution o firetree has likely not reached its potential extent in Hawai'i.

P Firetree infestations in leeward areas such as Kula on Maui and North

Kona, Hawai'i, extend up to 6,400 ft (1,940 m) and 6,040 ft (1,830 m) elevation. These records imply that infestations on Kapapala and Keauhou ranches and Hawaii Volcanoes National Park (in Ka'u District), with up er limits of 3,760 ft (1,140 m) and 4,000 ft (1,210 m), have not yet reac l! ed their potential upper elevational extent. Similarly, firetree is found down to 2,000 ft (600 m) and lower in protected sites in Hamakua and Hawaii Volcanoes National Park on'Hawai'i Island, on Liina'i, and on Kaua'i. Infestations along Hi hway 11 near Volcano Village and on Kapapala Ranch may not have reache f their potential lower elevation extent.

The rate at which firetree can spread into suitable habitats appears to be very rapid. A single individual was noted in Hawaii Volcanoes National Park between Kilauea Military Camp and Hawaiian Volcano Observatory by F.R. Fosberg in 1961 (Doty and Mueller-Dombois 1966). By 1978,1,304 a (609 ha) in the Park were mapped as su porting firetree infestations of various P 1978). Data rom the present study show 30,134 a

with firetree in the Park, including a 704-a (285-ha) increase during: the two-year course of the study. These records indicate a 20-fold increase in infested area within Hawaii Volcanoes in approximately eight years. Firetree has since read into suitable habitats that were not occupied at the time of this survey (1983-1985).

The distribution of firetree on Kaua'i (Fig. 5) includes infestations on the northwestern slopes of the Island that are noncontiguous with the center of distribution m the Koke6e/Waimea Canyon area. Separating these areas of firetree invasion is an area of closed-canopy mesic native forest. This pattern of invasion may indicate that the tree is limited in its ability to invade intact closed-canopy native forests. If so, it is possible that firetree is somewhat shade intolerant, and that its spread may be partially controlled by excluding disturbances, such as feral

U%iteaker and Gardner \ Firetree Distibution in Hawai'i 235

animals and forest fragmentation, in areas that support a closed canopy forest.

More enerally, the distribution of firetree throughout the State seems associate f with disturbance. Major areas of infestation are pastures, roadsides, trails, secondary forest, and steep, unstable slopes. In fact, the aggressive nature of the species was first noted in pasturelands. Even the areas of its distribution that are within Hawaii Volcanoes National Park have been subjected to fire, volcanic activity, and foragin by feral oats (Capra hircus), pigs, and cattle (Bos taurus). A f ditionally,

kretree has been identified as an actinorrhizal nitro en-fixing species i (Miguel and Rodriguez-Barrueco 1974; Turner and Vitouse 1988). Thus, the wdespread invasions observed in Hawai'i may be due to a competitive advantage gained throu~h nitrogen fixation on sites undergoing prima or secondary plant succession, especially on nitrogen-poor substrates s u x as those occurring in Hawaii Volcanoes National Park. This advantage may be lost on sites that have not been recently disturbed. However, in a landscape that is predominately influenced directly and/or indirectly by human activities, the further spread of firetree seems inevitable.

The com lex of environmental factors in a particular area may affect the otential e f evational range of firetree, even on a single island. In

Peeward areas such as North Kona, Hawai'i, and Kula, Maui, the distribution is between 3,200 and 6,400 ft (970-1,940 m). However, in windward areas such as Hamiha, Hawai'i, the distribution is between 2,000 and 4,400 ft (600-1,335 m) elevation. This implies that the species may be adapted to a lower elevational range in windward (wetter) areas than in leeward (drier areas in Hawai'i. In regions that are intermediate along rainfa1 gradients, firetree may have an intermediate potential elevational range.

I'

Phenology Flowering, immature fruiting, mature fruiting, and fruit drop seem to

follow an annual cycle that is endo enously controlled and/or initiated by B consistent annual environmental cyc es, such as day length. The timing of these phases in Hawai'i seems to be similar to that observed in the native habitat of the species (Krauss 1964; Gardner 1984). Fruiting was heaviest at the warmest and driest site and lightest at the coolest and wettest site. This suggests that firetree may not be well ada ted to reproducin in cold and/or extremely wet habitats. Also, our data ( \Rh iteaker and Gar & er 1987) show that while these phenophases have definite maximum and minimum activity periods, at no time during the course of this study were any of these phenophases competely absent.

Leaf flushin may also have a generally consistent annual cycle, as illustrated by t E e similar patterns at two of the study sites, and by the peak of activity occurrin in June of each year at all three sites. However, when the sligh tf y different overall pattern of leaf flushing at the driest site is compared with precipitation, maximum flushing seems to follow periods of high precipitation; minimum flushing seems to follow periods of low precipitation (Whiteaker and Gardner 1987). Thus, flushing of firetree may be sensitive to variations in precipitation where total

Whiteaker and Gardner \ Firetree Distribution in Hawai'i 236

annual precipitation is relatively low. Flushing was only absent at one site for one observation. Thus, although the amount of flushing may vary, this phenophase of firetree is also almost never absent.

The variation among sites in the pattern of the amount of leaf fall is shown by large peaks of leaf fall occurring at different times of the year at one study site relative to the other two sites. This may be due to the geographical osition of the sites relative to the volcanically active Pu'u '0'0 vent an 8 the corresponding incidence of high sulfur dioxide levels in the atmosphere. Thus, the large peaks in the amount of leaf fall may reflect the response of firetree to high sulfur dioxide concentrations superimposed on the otherwise steady rate of leaf fall of an evergreen species.

MANAGEMENT CONSIDERATIONS

Most ecologists familiar with the problem think that, if no action is taken, the present trend of rapid spread of firetree into native habitats will continue, resulting eventually in formation of dense, closed-canopy stands, beneath which native plant life will be excluded. This, in turn, will result in the destruction of habitats for native birds, including endangered species. However, sufficient funds and manpower are not available, nor is it anticipated that they will be provided in the future, for large-scale control of firetree through the labor-intensive chemical or mechanical removal approaches used in the ast. Such continued efforts would serve only to divert limited funding an a manpower away from efforts to control other alien p l a ~ t species for which these approaches may still be feasible.

Although chemical and/or mechanical control over large areas would be impractical, these approaches may be within the manpower and funding capability of management on comparatively small areas that contain particularly valuable resources, such as rare and endangered endemic species and the native plant communities that support them. In conjunction with this more focused a proach to chemical and/or mechanical control, biological control is the o d' y long-term hope for widespread control.

By promoting awareness of the threat of firetree to native ecosystems and ranchlands, the support of public land managers and ranchers adjacent to conservation areas can be obtained. This would aid in preventing firetree from being continually reintroduced into rotected areas. It

increased possibility of success. P would also enable managers to consider direct contro in larger areas with

The Hawaii Department of A riculture and the Department of Land and Natural Resources should again f eclare firetree a noxious weed and increase efforts toward its control on a statewide basis on State as well as rivately owned land. Support for biocontrol research for firetree should

ge encouraged by roviding funds for foreign exploration and related work. !i Federal agencies s ould cooperate in any State-supported control pro am and could provide biological control quarantine facilities as w e f as

Whiteaker and Gardner \ Firetree Distibution in Hawai'i 237

services of federally funded personnel. A cooperative Statewide control program would enhance the ability of management to control firetree and to

revent this species from becomng established in rotected areas such as h aleakala National Park. Public awareness wou f d then be much more widespread than at present, and public attitudes toward permitting the occurrence of ap arently innocuous infestations of firetree on private land could be alterecf with increased understanding of the need for control. The public should be less tolerant of firetree infestations on State and federal public lands than it is.

ACKNOWLEDGMENTS

We would like to express our appreciation to S. Anderson, P. Connally, L. Cuddihy, R. Daehler, R. Hobd , R. Karni, T. Lindsey, J. Obata, D. Sugawa, r E. Tamura, and T. Watanabe or their knowledge, advice, and logdcal support. Thanks to J. Dybdal and the Estate of James Campbell for access to the Palikea area on O'ahu. Gratitude is also due to A. de la Cruz and the Koke'e Museum, and H. Voris for lodging and logistical support. Thanks also to Mr. and Mrs. Eldon English for their permission to monitor trees on their land and for use of the precipitation data that they maintain.

This study was supported by funds made available through the University of Hawaii Cooperative National Park Resources Studies Unit, by National Park Service Cooperative Agreement CA 8004-2-0001.

Whiteaker and Gardner \ Firetree Distribution in Hawai'i

Literature Cited

Anonymous. 1962. Noxious weeds of Hawaii. Hawaii Dep. Agric. Div. Plant Industry. Mimeo.

Clarke, G. 1978. The distribution of Myrica f q a and other selected problem exotics within Hawaii Volcanoes National Park. Pres. 2nd Conf. Nat. Sci., Hawaii Volcanoes Natl. Park, June 1-3, 1978. Abstract, Proc. 2nd Conf. Nut. Sci., Hawaii Volcanoes Natl. Park, 51. Honolulu: Univ. Hawaii Coop. Natl. Park Resour. Stud. Unit.

Doty, M.S., and D. Mueller-Dombois. 1%6. Atlas for bioecology shrdies in Hawaii Volcanoes National Park. Hawaii Bot. Sci. Paper No. 2. Botany Dep. Honolulu: Univ. Hawaii.

Dunevitz, V.L., and JJ. Ewel. 1981. Melopathy of wax myrtle (Myrica cerifera) on Schinus terebinthifolius. Flu. Scientist 44:l-13.

Ewel, JJ., D.S. Ojima, DA. Karl, and W.F. DeBusk. 1982. Schinus in successional ecosystems of Eveq$ades National Park. Tech. Rep. T-676, Natl. Park Sew. So. Ha. Res. Cent., Everglades Natl. Park. Homestead, Ha.

Foote, D.E., E.L. Hill, S. Nakamura, and F. Stephens. 1972. Soil survey of the islands of Kauai, Oahu, Maui, Molokai, and Lanai, State of Hawaii. Honolulu: U.S. Dep. Agric., Soil Conservation Service, in cooperation with Univ. Hawaii Agric. Exp. Sta.

Fosberg, F.R. 1937. Immigrant plants in the Hawaiian Islands. I. Univ. Hawaii Occas. Paper No. 32. Honolulu: Univ. Hawaii.

Gardner, D.E. 1984. Current biocontrol investigations: exploration for control agents in the native habitats of firetree, and evaluation of rust diseases of Rubus spp. in the southeastern U.S. Proc. Fijlh Conf. Nut. Sci., Hawaii Volcanoes Natl. Park, 53-60. Honolulu: Univ. Hawaii Coop. Natl. Park Resour. Stud. Unit.

Gardner, D.E. 1985. Observations on some unusual flowering characteristics of Myrica f q a . Newsletter, Ha wn . Bot. Soc. 24: 14- 17.

Gardner, D.E., and C.J. Davis. 1982. The prospects for biological control of nonnative plants in Hawaiian national parks. Tech. Rep. 45, Univ. Hawaii Coop. Natl. Park Resour. Stud. Unit. Honolulu.

Gardner, D.E., and VA.D. Kageler. 1982. Herbicidal control of jiretree in Hawaii Volcanoes National Park: a new approach. Natl. Park Sen. Ecol. Sen. Bull. 7. Washington D.C.: U.S. Dep. Interior.

Haselwood, E.L., and G.G. Motter, eds. 1983. Handbook of Hawaiian weeds, 2nd ed., revised and expanded by R.T. Hiano. Harold L. Lyon Arboretum. Honolulu: Univ. Hawaii Pr.

Hodges, C.S., Jr., and D.E. Gardner. 1985. Myrica fay= potential biological control agents. Tech. Rep. 54, Univ. Hawaii Coop. Natl. Park Resour. Stud. Unit. Honolulu.

Whiteaker and Gardner \ Firetree Distribution in HawaiT 239

Hosaka, E.Y. 1945. Noxious weeds of Hawaii. Honolulu: Board of Commissioners of Agriculture and Forestry, in cooperation with Univ. Hawaii Agric. Exp. Sta.

Hosaka, E.Y., and A. Thistle. 1954. Noxious plants of the Hawaiian ranges. Univ. Hawaii Agric. Ext. Serv. Bull. 62. Honolulu: Univ. Hawaii.

Krauss, N.L.H. 1964. Insects associated with fuebush (Myrica faya Aiton). Proc. H a m . Entomol. SOC. 18(3):405-411.

LaRosa, A.M., C.W. Smith, and D.E. Gardiler. 1985. Role of alien and native birds in the dissemination of fuetree (Myrica faya Ait. - Myricaceae) and associated plants in Hawaii. Pac. Sci. 39:372-378.

Lawrence, G.H.M. 1951. Taxonomy of vascularplants. New York: The MacMillan Co.

Miguel, C., and C. Rodriguez-Barrueco. 1974. Acetylene-reducing activity of detached root nodules of Myrica faya Ait. Plant and Soil 41521-526.

Mueller-Dombois, D., and F.R. Fosberg. 1974. Vegetation map of Hawaii Volcanoes National Park. Tech. Rep. 4, Univ. Hawaii Coop. Natl. Park Resour. Stud. Unit. Honolulu.

Neal, M.C. 1965. In gardens of Hawaii. B.P. Bishop Mus. Spec. Pub. 50. Honolulu: Bishop Mus. Pr.

Ripperton, J.C., and E.Y. Hosaka. 1942. Vegetation zones of Hawaii. Hawaii Agric. Exp. Sta. Bull. 89. Honolulu: Univ. Hawaii.

Skolmen, R.G. 1979. Plantings on the forest reserves of Hawaii 1910-1960. Inst. Pac. Islands For., Pac. SW For. Range Exp. Sta. USDA For. Serv., Honolulu.

Smathers, GA., and D.E. Gardner. 1979. Stand analysis of an invading fuetree (Myrica faya Aiton) population, Hawaii. Pac. Sci. 33(3):239-255.

Smith, C.W. 1985. Impact of alien plants on Hawai'i's native biota. In Hawai'i's terreshial ecosystems: preservation and management, ed. C.P. Stone and J.M. Scott, 180-250. Univ. Hawaii Coop. Natl. Park Resow. Stud. Unit. Honolulu: Univ. Hawaii Pr.

Stone, C.P., and D.D. Taylor. 1984. Status of feral pig management and research in Hawaii Volcanoes National Park. Proc. 5th Con& Nut. Sci., Hawaii Volcanoes Natl. Park, 106-117. Honolulu: Univ. Hawaii Coop. Natl. Park Resow. Stud. Unit.

Turner, D.R., and P.M. Vitousek. 1988. Nodule biomass of the nitrogen-fhg alien Myrica faya Ait. in Hawaii Volcanoes National Park. Pac. Sci. 41(1-4):186-190.

Walters, GA., and W.S. Null. 1970. Controlling firetree in Hawaii by injection of Tordon 22K. Research Note PSW-217. Berkeley: USDA For. Serv.

Watanabe, T. 1982. Noxious weeds of Hawaii. Hawaii Dept. Agric. Div. Plant Industry. Unpub. report.

Whiteaker and Gardner \ Firetree Distibution in Hawai'i 240

Whiteaker, L.D., and D.E. Gardner. 1985. The distibution of Myrica faya Ait. in the state of Hawai'i. Tech. Rep. 55, Univ. Hawaii Coop. Natl. Park Resour. Stud. Unit. Honolulu.

Whiteaker, L.D., and D.E. Gardner. 1987. The phenology and stand shucture of Myrica faya Ait. in Hawaii. Tech. Rep. 62, Univ. Hawaii Coop. Natl. Park Resour. Stud. Unit. Honolulu.

Yamayoshi, H.T. 1954. M. f a y survey of Paauilo and Kukaiau Ranch. Plant Quarantine Inspector, Territorial Agric. Dep. Rep. Unpub.